416 ELECTRO-PHYSIOLOGY CHAP. 



the lateral surfaces of the organ. They are directed inwards as 

 well as outwards through the body of the fish, and further fill up 

 the cavity." 



With regard to the immunity of Torpedo to its own shocks, 

 it is to be noted that " the currents which flow along the back, 

 from the median walls of the organ to the middle line, and con- 

 versely on the ventral surface from this line to the border, 

 necessarily find their path through brain and cord ; and, since 

 this is the shortest path between the most active parts of the 

 two organs, there can be no stronger current through the torpedo " 

 (du Bois-Eeymond). 



Du Bois-Eeymond was able to reproduce all these effects 

 upon artificial models, by grouping zinc -platinum elements 

 together in series like electrical plates, and dipping them sud- 

 denly into water, upon which the current spread itself as in a 

 discharge, and was led off in a similar manner. 



On leading off from symmetrical points of the dorsal or 

 ventral surface, du Bois-Eeymond obtained deflections from his 

 fish during the shock, which might be due to unequal innervation 

 of the two organs. 



In Malapterurus du Bois-Eeymond established that " during 

 the discharge each point of the organ proximal to the tail is 

 positive to those nearer the head, and that it is unimportant 

 whether the point lies at the circumference of a given cross- section 

 of the fish, or at its back, side, or belly," so that the polar surfaces 

 of the organ lie, as in G-ymnotus, towards the head and tail. 



It follows that the direction of the normal discharge in 

 electrical fishes is invariably at right angles to the plane of the 

 plates. In Torpedo, where the plates are horizontal, with normal 

 position of the animal, the discharge therefore occurs between 

 back and belly ; in Gymnotus, on the contrary, where the plates 

 are, as a rule, vertical to the long axis of the organ, i.e. in the 

 transverse plane of the animal, the discharge passes longitudinally 

 from head to tail. The same is the case in Malapterurus, where 

 the plates exhibit a similar arrangement. 



A very remarkable rule at first appeared to be indicated in the 

 fact, as pointed out by Pacini, that the distribution of the nerves in 

 Torpedo and Gymnotus occurs always upon that surface of the plate 

 which is negative in the discharge, i.e. the lower surface in Torpedo, 

 the posterior surface in Gymnotus. In this last animal Faraday 



